A Washington County judge on Tuesday will sentence Robert Atrops for the three-decades-old murder of his wife, a crime that comes with a mandatory sentence of life in prison.

Atrops was convicted in April, after a nearly three-week trial in which jurors heard evidence they found compelling enough to vote to convict the 70-year-old after just six hours of deliberations. Much of the evidence they heard is considered circumstantial — there were no witnesses to the crime — and the trial demonstrated both the challenges of prosecuting a decades-old cold case and the limitations and potential pitfalls of forensic evidence.

Deborah Atrops was found strangled on Dec. 1, 1988, her body stuffed into the trunk of her car near Beaverton. The case went unsolved until it was reopened by Washington County investigators decades later. DNA evidence, in the form of skin cells found on Deborah Atrops’ coat in 2022, led to Robert Atrops’ arrest for the murder.

In the final argument jurors heard during the trial’s close in April, prosecutors laid out their theory about what that forensic evidence meant.

“We know that his DNA was found on her neck … on her collar, and other area of her coat. No one else’s was, right?” Senior Deputy District Attorney Allison Brown told jurors. “They didn’t actually swab that jacket until a couple years ago. So after 35 years, there’s still the defendant‘s DNA on her jacket.”

“We’ve got evidence of actually having that physical contact with her that‘s directly connected to her murder.”

That contradicts what DNA analysts from the Oregon State Police Crime Lab, who worked on the case with Brown, testified just days earlier. They told jurors there’s no way to know when the DNA got on the coat. And they said they couldn’t tell how it got there. Before Brown’s closing, no one from the lab said they’d tested Deborah’s neck for DNA.

When OPB asked Brown if she was concerned about her statements being inconsistent with what analysts from the Oregon crime lab testified, Brown demurred.

“The lab’s job isn’t to advocate or to make any conclusions,” she said. “That‘s our job, to do our best to piece it together for the jury and make our arguments.”

One of Atrops’ attorneys, Janis Puracal, said in a statement that DNA being misinterpreted in court was what they worried about all along.

“The State used unreliable DNA evidence. This isn’t the DNA you see on television,” she said. “The State just convicted an innocent man.”

Since the start, the DNA evidence in the Atrops case posed a challenge for both sides. According to the Oregon lab’s own internal charts, the DNA results were relatively weak, and the finding is based on complex statistics, far from how DNA is depicted in popular culture.

It‘s the task of both the prosecution and defense to present a compelling story to the jury, a narrative of events that draws on what was presented during the trial. What Brown’s closing remarks demonstrate is that the final words jurors hear can stretch the bounds of what the evidence shows. There’s a different threshold for accuracy because those statements aren’t evidence; they’re an argument.

The difficulty of DNA evidence

Of all the evidence presented at trial, modern DNA analysis may be some of the most difficult to explain to a jury so that it‘s simple enough to understand and also still scientifically accurate. Since it first revolutionized criminal cases some 40 years ago, DNA testing has become far more advanced. Tests are increasingly sensitive. A swab taken from a crime scene may not only contain the DNA of a victim and suspect, but also anyone else who came into contact with or stepped into the same space where a crime took place.

With so many possible DNA contributors picked up in a test, the results can be harder for lab analysts to interpret and ultimately determine what DNA belongs to whom. To figure that out, crime labs the world over turn to “probabilistic genotyping.” These are software programs that use algorithms to decipher DNA evidence when more than one person is involved. The algorithms calculate the likelihood of observing a DNA sample if the suspect‘s DNA was included, and compare that to the likelihood that a random person’s DNA was in the mixture instead. A software program spits out that number, which lab techs typically then help prosecutors decipher.

“It‘s very powerful, but it‘s extremely complicated and it‘s hard for people to get their heads around it,” said William Thompson, professor emeritus at UC Irvine who specializes in forensic science.

In the case of Robert Atrops, the Oregon State Police Crime Lab conducted three DNA tests on the coat Deborah Atrops wore when she was killed. Defense attorneys have challenged the process that the analysts at the lab followed. The lab’s first two tests essentially ruled him out. The third time around, after adjusting the test parameters, analysts got a 132. That number means it‘s 132 times more likely to see the DNA mixture they found on the coat if Robert Atrops was a DNA contributor, as opposed to a random person. That number is one of the weakest the lab could have gotten, barely above an “inconclusive” result, according to its internal scale.

This kind of result complicates the common understanding of forensic science. In movies and shows, DNA tests often result in “matches” — irrefutable statements about whose genetic material is present in a test. But DNA tests are always based on probability or likelihood, especially if they involve more than one person.

“We don’t know the ground truth for any forensic sample,” Heather Feaman, the DNA analyst who worked on the Atrops case, said during a pretrial hearing last October. “Unless you have made that sample, you don’t know what the ground truth is.”

She also noted the DNA results could not be called a “match.”

During cross-examination by Atrops’ attorneys at trial, Feaman told the jury that the DNA finding by itself communicated very little.

“You cannot tell us when any of the DNA got on the jacket,” defense attorney Puracal pressed Feaman about the limitations of the lab’s tests.

“No, DNA can’t tell you that,” Feaman responded.

“You cannot tell us if any of the DNA got on the jacket days or weeks or months before the murder,” Puracal continued.

“Nope, cannot tell you that,” Feaman replied.

Even though these limitations were made plain in court, studies show jurors often bring preconceived notions of what DNA evidence means.

“They think of DNA evidence as being the most powerful and persuasive kind of evidence there is,” Thompson, a former prosecutor, said. “That‘s a potential kind of prejudice.”

This skewed perception of DNA’s strength, regardless of what the evidence shows, is one reason some attorneys believe there needs to be more scrutiny for when and how it gets used in court.

Bess Stiffelman, a California-based attorney and consultant who trains lawyers in how to understand forensic evidence, said the problem with using probabilistic genotyping software in cases like these — where a sample could contain multiple people’s DNA or the result the software gives is a low number — is that people have trouble understanding probability or the overall limitations of DNA evidence. And those misunderstandings can skew a case.

“The danger just seems to really outweigh the benefit here, unless the benefit is just for the prosecution to get a conviction,” she said. “It doesn’t seem like this is leading to a fair hearing.”

Making an argument

Before the Atrops’ case began, defense attorneys attempted to get the DNA evidence excluded from the trial altogether. They argued the lab conducted the tests incorrectly.

Prosecutors, meanwhile, said the lab did nothing wrong. They noted that the lab regularly conducts DNA analysis for thousands of cases statewide, and thus, the state lab workers can be relied upon as experts.

A 1993 U.S Supreme Court ruling requires trial court judges to determine if scientific evidence is reliable enough to bring to trial. Two years later, the Oregon Supreme Court brought that same standard into state court. The safeguards are meant to keep inaccurate or pseudoscience from coming before a jury as evidence. The decision to allow the DNA evidence in was left to Washington County Circuit Court Judge Oscar Garcia.

Garcia determined that in the Atrops case, it would be up to the attorneys to argue the significance of the DNA evidence — with prosecutors arguing its reliability, and the defense pointing out its limitations.

The judge also warned that the way the DNA is discussed before jurors could be unfair.

“What exactly can be said to the jury about this — the language?” Garcia asked. “It‘s not a ‘match.’ They can’t use that language. So what exactly is the language that we’re talking about here?”

“The DNA people will not say it‘s a match,” Washington County Deputy District Attorney Chris Lewman told Garcia. “Ms. Brown and I, during opening and closing, we’re not gonna say it‘s a match because you’re right, that‘s not what we’re saying.”

During closing arguments, Lewman had a difficult task: trying to stick to what the judge said and make a compelling argument at the same time.

“Robert Atrops cannot be excluded as a contributor,” he stated. “I’m not telling you that it is a match, right? We can’t say that.”

How jurors interpret what lawyers are, or aren’t, saying can depend on what they already assume about forensic evidence, according to all three legal experts OPB interviewed.

“It‘s a nod and a wink kind of thing,” Thompson said. “‘We’re not supposed to say match, but you know what we mean, right?’ That capitalizes on people’s background understanding.”

Closing arguments can be some of the most pivotal moments during a trial. It‘s the last word the jury hears before deliberating, and can have an outsized impact on how verdicts are determined.

They also don’t hold the same evidentiary standards as what is presented at trial. In the instructions given to members of the jury in the Atrops case, the judge told jurors to base their decision on the evidence, noting that “lawyers’ statements and arguments are not evidence.”

That‘s because in these statements, attorneys can lean on inferences — drawing conclusions based on the available evidence and reasonable logic — rather than sticking to technical language or strict facts.

In an interview with OPB after the verdict, Brown said that‘s what she was presenting to the jury during closing — inferences.

“When we’re looking at everything and making our argument, we’re not just taking that piece, we’re taking all the other information too,” she said. “So we’re saying ‘look at all the other information we have as well, there’s evidence that he’s the one that did this to her,’ and to tie all that together for them.”

It‘s unclear how much weight jurors placed on either side’s closing statements in their conviction of Robert Atrops. But Daniel Medwed, professor of criminal justice at Northeastern University, said these arguments provide a narrative path for jurors to justify their own decision-making.

“They were basically giving the jury a hook to hang their hat on if they felt like he did it,” he said. “The jury could say, ‘Well, there’s science that supports the theory that he was there.’ So you can see why the prosecutors would want it in the case, warts and all.”

A case that Washington County could find itself litigating again. Robert Atrops’ attorneys plan to file an appeal, challenging his conviction.

News Source : https://www.opb.org/article/2025/05/20/washington-county-dna-evidence-murder-robert-deborah-atrops-cold-case-evidence-crime/